MANAGEMENT OF DATA CENTER SDN BASED ON THE MODIFIED DEVOFLOW APPROACH
DOI:
https://doi.org/10.17721/AIT.2025.1.05Keywords:
Software-defined Networking, data center, OpenFlow, Hedera, DevoFlow, elephant-flows, mice-flows, SDN controller, OpenFlow Switch, flow table.Abstract
Background. The paper investigates the features of implementing the SDN network management layer when servicing data center flows of different volumes based on Hedera technology, the OpenFlow protocol, and the DevoFlow approach. The distribution of processing functions, collection of statistical data, and the frequency of updating statistical data affect both the amount of service information and the efficiency and effectiveness of management in terms of distributing incoming traffic for load balancing. The aim of the work is to modify the Hedera approach by redistributing the functions of flow classification and adaptive control depending on the flow volume at the control level, as well as to develop an algorithm for determining the allowable time for sending statistics from switches to the SDN controller
Methods. The method of system analysis and decomposition is used to study complex systems, methods of collecting data on the state of the network, as well as heuristic rules for determining the volume of service traffic.
Results. The paper analyzes the features of construction and operation of technologies like Hedera and DevoFlow, recommend a modified architecture for building the control plane and identifies the main functions of software modules that can improve the efficiency of SDN controllers. Due to changing the procedure for collecting and processing service information on the equipment of the SDN network of data centers and returning to a centralized type of control.
Approaches to the classification of incoming flows and their division into small “mice-flows”, medium “medium-flows” and large “elephant-flows” are determined, which allows further use of multi-path routing for more efficient use of network resources. Also, this article proposes an algorithm for determining the permissible time for sending statistics from switches (push-based method), which takes into account the dynamics of load changes in the lines in the network, the number of active sessions and is based on a heuristic rule for limiting the amount of service traffic.
Conclusions. A modified architecture of the control level based on the DevoFlow approach has been developed and the main functions of the software modules of the SDN controller of the data center network have been determined. Also, some analytical dependencies have been developed to determine the approved time for sending statistics from switches to the SDN controller, taking into account the current state of network congestion. Additionally, specific experiments were conducted to confirm both the reliability of the chosen method for obtaining statistics and the proposed algorithm for determining the permissible time for sending statistics from switches to the SDN controller.
Downloads
References
Costa, L. C., Vieira, A. B., de Britto e Silva, E., Macedo, D. F., Vieira, L. F. M., Vieira, M. A. M., da Rocha Miranda, M., Batista, G. F., Polizer, A. H., Gonçalves, A. V. G. S., Gomes, G., & Correia, L. H. A. (2021). OpenFlow data planes performance evaluation. Performance Evaluation, 147. https://doi.org/10.1016/j.peva.2021.102194
Gilliard, E., Liu, J., Aliyu, A. A., Juan, D., Jing, H., & Wang, M. (2024). Intelligent load balancing in data center software‐defined networks. Transactions on Emerging Telecommunications Technologies, 35(4). https://doi.org/10.1002/ett.4967
Globa, L., Skulysh, M., Romanov, O., & Nesterenko, M. (2019). Quality Control for Mobile Communication Management Services in Hybrid Environment. In Ilchenko, M., Uryvsky, L., Globa, L. (Eds.), Advances in Information and Communication Technologies: Vol. 560. Lecture Notes in Electrical Engineering
(pp. 76–100). Springer. https://doi.org/10.1007/978-3-030-16770-7_4
Romanov, O., & Mankivskyi, V. (2019). Optimal Traffic Distribution Based on the Sectoral Model of Loading Network Elements. In 2019 IEEE International Scientific-Practical Conference Problems of Infocommunications, Science and Technology (PIC S&T) (pp. 683–688). IEEE. https://doi.org/10.1109/picst47496.2019.9061296
Romanov, O., Nesterenko, M., & Mankivskyi, V. (2021). The method of redistributing traffic in mobile network. In D. Ageyev, T. Radivilova, & N. Kryvinska (Eds.), Data-Centric Business and Applications (Lecture Notes on Data Engineering and Communications Technologies, Vol. 69, pp. 159–182). Springer. https://doi.org/10.1007/978-3-030-71892-3_7
Romanov, O., Siemen, E., Nesterenko, M., & Mankivskyi, V. (2021). Mathematical Description of Control Problems in SDN Networks. In International Conference on Applied Innovations in IT (ICAIIT) (pp. 33–40). http://dx.doi.org/10.25673/36582
Rout, S., Patra, S.S., Sahoo, B. (2017). Performance Evaluation of the Controller in Software-Defined Networking. У H. Behera, & D. Mohapatra (Eds.) Computational Intelligence in Data Mining: Vol 556. Advances in Intelligent Systems and Computing (с. 543–551). Springer.
https://doi.org/10.1007/978-981-10-3874-7_51
Shirmarz, A., & Ghaffari, A. (2020). Performance issues and solutions in SDN-based data center: a survey. The Journal of Supercomputing, 76(10),
7545–7593. https://doi.org/10.1007/s11227-020-03180-7
Tennakoon, D., Chowdhury, M., & Luan, T. H. (2018). Cloud-based load balancing using double Q-learning for improved Quality of Service. Wireless Netw, 29, 1043–1050. https://doi.org/10.1007/s11276-018-1888- 8
Wang, C., Cao, W., Hu, Y., & Liu, J. (2023). Data Center Traffic Scheduling Strategy for Minimization Congestion and Quality of Service Guaranteeing. Computers, Materials & Continua, 75(2), 4377–4393. https://doi.org/10.32604/cmc.2023.037625
Yu, H., Qi, H., Li, K., Zhang, J., Xiao, P., & Wang, X. (2018). Openflow Based Dynamic Flow Scheduling with Multipath for Data Center Networks. Computer Systems Science and Engineering, 33(4), 251–258. https://doi.org/10.32604/csse.2018.33.251
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Copyright (c) 2025 Advanced Information Technology
This work is licensed under a Creative Commons Attribution 4.0 International License
